When network protection is implemented in a cross-connect switch, there are stringent requirements concerning the amount of time needed to perform the protection switch once a problem has been detected. Typically, the protection switch is performed by cutting another connection through the matrix of the cross-connect switch. However, cutting another connection through the matrix has various disadvantages. Cross-connection algorithms generally take more time to complete than is allowed by the protection switching requirements. For example, the protection switching requirements specify tens of milliseconds rather than the hundreds of milliseconds it takes to cut a new connection. Creating and deleting cross-connections may not meet the desired switching times. If center stage rearrangements are required in a standard three stage non-blocking matrix, then the time needed to adjust matrix connections for a protection switch may take even longer. Therefore, it is desirable to avoid the slower than required changes in matrix connections to perform a protection switch in a cross-connect switch.